The overall goal of this proposal is to identify the molecular mechanisms underlying cell competition, in which normally viable cells are eliminated via apoptosis when they grow next to cells with higher growth rates. During cell competition, the loss of the "loser" cells is compensated by extra proliferation of the "winners", thus the organ develops to a normal size with appropriate cell fates. Our previous work demonstrated that cells with different expression levels of the c-myc proto-oncogene ortholog dMyc compete with each other, and that this competition has a critical role in regulating Drosophila wing size. What makes some cells "winners" of the competition and other cells "losers"? Based on our previous work, we hypothesize that dMyc-induced cell competition is mediated by a short-range signal that leads to specific genetic programs in both the "winners" and the "losers". With this in mind, we propose a systematic approach to identify the molecular mechanisms underlying cell competition. Using both an in vivo wing disc model system and a cell-culture based cell competition assay, the experiments we propose here will address how "winner" and "loser" cells are determined, and how information about their status is communicated between the cell populations. Our primary objectives are to determine the gene expression differences between competing cell populations and obtain a genetic signature of cell competition, to identify genes that are required for cell competition to occur, and to address the function of these genes in the competitive process. Since cell competition has also been documented in the mouse, the mechanism of cell competition as well as its role in development is likely conserved between flies and vertebrates. Furthermore, Myc expression is increased in many cancers, thus such a mechanism could also be used by incipient tumors to kill off nearby normal cells and expand their territory. We postulate that cell competition in Drosophila may provide a unique model with which genes that are involved in the earliest steps of cancer progression can be identified. [unreadable] [unreadable] [unreadable]